diff --git a/tests/test_script_split_integrity.py b/tests/test_script_split_integrity.py
new file mode 100644
index 00000000..89a12e5a
--- /dev/null
+++ b/tests/test_script_split_integrity.py
@@ -0,0 +1,304 @@
+"""Tests for the script.js → module split integrity.
+
+Verifies that:
+ - The monolithic script.js no longer exists
+ - All expected split modules are present on disk
+ - index.html loads all split modules via <script> tags
+ - core.js loads first and init.js loads last (ordering contract)
+ - No duplicate top-level function declarations across modules
+ - Every onclick="fn(…)" in index.html has a matching function
+   declaration in one of the split modules
+ - No module references undefined globals at parse time via
+   window.X = X assignments (the bug pattern we fixed)
+"""
+
+import os
+import re
+from pathlib import Path
+from collections import defaultdict
+
+import pytest
+
+# ---------------------------------------------------------------------------
+# Paths
+# ---------------------------------------------------------------------------
+
+_ROOT = Path(__file__).resolve().parent.parent
+_STATIC = _ROOT / "webui" / "static"
+_INDEX = _ROOT / "webui" / "index.html"
+
+# The 17 modules that replaced script.js (order matters for first/last checks)
+SPLIT_MODULES = [
+    "core.js",
+    "media-player.js",
+    "settings.js",
+    "search.js",
+    "sync-spotify.js",
+    "downloads.js",
+    "wishlist-tools.js",
+    "sync-services.js",
+    "artists.js",
+    "api-monitor.js",
+    "library.js",
+    "beatport-ui.js",
+    "discover.js",
+    "enrichment.js",
+    "stats-automations.js",
+    "pages-extra.js",
+    "init.js",
+]
+
+# Other JS files that exist in static/ but are NOT part of the split
+NON_SPLIT_JS = {"setup-wizard.js", "docs.js", "helper.js", "particles.js", "worker-orbs.js"}
+
+# Pre-existing duplicate helper functions that lived in the original monolith.
+# In a plain <script> context the last-loaded declaration wins.  These are NOT
+# regressions from the split — they should be deduplicated in a follow-up.
+KNOWN_CROSS_FILE_DUPES = {
+    "escapeHtml",        # downloads.js, artists.js, discover.js
+    "formatDuration",    # sync-spotify.js, wishlist-tools.js, sync-services.js
+    "matchedDownloadTrack",    # downloads.js, wishlist-tools.js
+    "matchedDownloadAlbum",    # downloads.js, wishlist-tools.js
+    "matchedDownloadAlbumTrack",  # downloads.js, wishlist-tools.js
+    "_esc",              # library.js, stats-automations.js
+    "_escAttr",          # downloads.js, stats-automations.js
+    "_formatDuration",   # stats-automations.js, pages-extra.js
+    "loadDashboardData", # search.js, wishlist-tools.js
+}
+
+# Pre-existing same-file duplicates (two filter UIs reuse the same names).
+KNOWN_SAME_FILE_DUPES = {
+    "applyFiltersAndSort",
+    "calculateRelevanceScore",
+    "handleFilterClick",
+    "initializeFilters",
+    "resetFilters",
+}
+
+# ---------------------------------------------------------------------------
+# Helpers
+# ---------------------------------------------------------------------------
+
+_FUNC_DECL_RE = re.compile(r"^(?:async\s+)?function\s+([A-Za-z_$][A-Za-z0-9_$]*)\s*\(", re.MULTILINE)
+_ONCLICK_RE = re.compile(r'onclick="([^"]*)"')
+_ONCLICK_FN_RE = re.compile(r"^([A-Za-z_$][A-Za-z0-9_$]*)\s*\(")
+_WINDOW_ASSIGN_RE = re.compile(
+    r"^window\.([A-Za-z_$][A-Za-z0-9_$]*)\s*=\s*([A-Za-z_$][A-Za-z0-9_$]*)\s*;",
+    re.MULTILINE,
+)
+_SCRIPT_SRC_RE = re.compile(r"filename='([^']+\.js)'")
+
+
+def _read(path: Path) -> str:
+    return path.read_text(encoding="utf-8", errors="replace")
+
+
+def _all_function_decls(js_text: str) -> list[str]:
+    """Return all top-level function declaration names in a JS file."""
+    return _FUNC_DECL_RE.findall(js_text)
+
+
+def _script_load_order(html: str) -> list[str]:
+    """Return the ordered list of JS filenames loaded from index.html."""
+    return _SCRIPT_SRC_RE.findall(html)
+
+
+# =========================================================================
+# Group A — File Existence
+# =========================================================================
+
+class TestFileExistence:
+    """The old monolith is gone and all split modules are present."""
+
+    def test_monolith_removed(self):
+        assert not (_STATIC / "script.js").exists(), "script.js should have been removed"
+
+    @pytest.mark.parametrize("module", SPLIT_MODULES)
+    def test_split_module_exists(self, module):
+        path = _STATIC / module
+        assert path.exists(), f"{module} missing from webui/static/"
+        assert path.stat().st_size > 0, f"{module} is empty"
+
+
+# =========================================================================
+# Group B — index.html Script Loading
+# =========================================================================
+
+class TestScriptLoading:
+    """index.html references every split module in the correct order."""
+
+    @pytest.fixture(autouse=True)
+    def _load_html(self):
+        self.html = _read(_INDEX)
+        self.loaded = _script_load_order(self.html)
+
+    @pytest.mark.parametrize("module", SPLIT_MODULES)
+    def test_module_loaded_in_html(self, module):
+        assert module in self.loaded, f"{module} not loaded in index.html"
+
+    def test_core_loads_first(self):
+        """core.js must be the first split module loaded."""
+        split_in_html = [f for f in self.loaded if f in SPLIT_MODULES]
+        assert split_in_html[0] == "core.js", (
+            f"Expected core.js first, got {split_in_html[0]}"
+        )
+
+    def test_init_loads_last(self):
+        """init.js must be the last split module loaded."""
+        split_in_html = [f for f in self.loaded if f in SPLIT_MODULES]
+        assert split_in_html[-1] == "init.js", (
+            f"Expected init.js last, got {split_in_html[-1]}"
+        )
+
+    def test_no_duplicate_script_tags(self):
+        """Each module should only be loaded once."""
+        split_in_html = [f for f in self.loaded if f in SPLIT_MODULES]
+        assert len(split_in_html) == len(set(split_in_html)), (
+            "Duplicate script tags detected"
+        )
+
+
+# =========================================================================
+# Group C — No Duplicate Function Declarations
+# =========================================================================
+
+class TestNoDuplicateFunctions:
+    """No two split modules should declare the same top-level function."""
+
+    @pytest.fixture(autouse=True)
+    def _scan_all(self):
+        self.func_map: dict[str, list[str]] = defaultdict(list)
+        for module in SPLIT_MODULES:
+            text = _read(_STATIC / module)
+            for fn_name in _all_function_decls(text):
+                self.func_map[fn_name].append(module)
+
+    def test_no_new_cross_file_duplicates(self):
+        """Catch NEW duplicate declarations; known pre-existing ones are allowed."""
+        dupes = {
+            fn: files
+            for fn, files in self.func_map.items()
+            if len(files) > 1
+            and fn not in KNOWN_CROSS_FILE_DUPES
+            and fn not in KNOWN_SAME_FILE_DUPES
+        }
+        assert not dupes, (
+            "NEW duplicate function declarations across modules:\n"
+            + "\n".join(f"  {fn}: {files}" for fn, files in sorted(dupes.items()))
+        )
+
+    def test_known_dupes_still_tracked(self):
+        """Ensure the known-dupe set stays current (remove entries when deduped)."""
+        actual_dupes = {fn for fn, files in self.func_map.items() if len(files) > 1}
+        stale = (KNOWN_CROSS_FILE_DUPES | KNOWN_SAME_FILE_DUPES) - actual_dupes
+        assert not stale, (
+            f"These known duplicates were resolved — remove from KNOWN_*_DUPES:\n"
+            + "\n".join(f"  {fn}" for fn in sorted(stale))
+        )
+
+
+# =========================================================================
+# Group D — onclick Handler Coverage
+# =========================================================================
+
+class TestOnclickCoverage:
+    """Every onclick="fn(…)" in index.html should have a matching
+    function declaration in the combined split modules."""
+
+    @pytest.fixture(autouse=True)
+    def _scan(self):
+        # Collect all function declarations from split modules
+        self.all_fns: set[str] = set()
+        for module in SPLIT_MODULES:
+            text = _read(_STATIC / module)
+            self.all_fns.update(_all_function_decls(text))
+
+        # Also include non-split JS files that are loaded
+        for extra in ("setup-wizard.js", "docs.js", "helper.js"):
+            path = _STATIC / extra
+            if path.exists():
+                self.all_fns.update(_all_function_decls(_read(path)))
+
+        # Extract all onclick function references from HTML
+        html = _read(_INDEX)
+        self.onclick_fns: set[str] = set()
+        for onclick_val in _ONCLICK_RE.findall(html):
+            m = _ONCLICK_FN_RE.match(onclick_val.strip())
+            if m:
+                fn_name = m.group(1)
+                # Skip JS keywords that happen to match (if, return, etc.)
+                if fn_name not in ("if", "return", "var", "let", "const", "this"):
+                    self.onclick_fns.add(fn_name)
+
+    def test_all_onclick_handlers_defined(self):
+        missing = self.onclick_fns - self.all_fns
+        assert not missing, (
+            f"onclick handlers reference undefined functions:\n"
+            + "\n".join(f"  {fn}" for fn in sorted(missing))
+        )
+
+    def test_onclick_count_sanity(self):
+        """Sanity check: there should be a substantial number of onclick handlers."""
+        assert len(self.onclick_fns) > 50, (
+            f"Only found {len(self.onclick_fns)} onclick handlers — expected 100+"
+        )
+
+
+# =========================================================================
+# Group E — No Dangerous Cross-File window.X = X Assignments
+# =========================================================================
+
+class TestNoCrossFileWindowAssignments:
+    """window.X = X at the top level of a module is only safe if X is
+    defined in that same module.  If X lives in a later-loading module,
+    this causes a ReferenceError at parse time."""
+
+    @pytest.fixture(autouse=True)
+    def _scan(self):
+        self.module_fns: dict[str, set[str]] = {}
+        self.window_assigns: dict[str, list[tuple[str, str]]] = defaultdict(list)
+
+        for module in SPLIT_MODULES:
+            text = _read(_STATIC / module)
+            self.module_fns[module] = set(_all_function_decls(text))
+            for prop, value in _WINDOW_ASSIGN_RE.findall(text):
+                self.window_assigns[module].append((prop, value))
+
+    def test_no_cross_file_references(self):
+        bad = []
+        for module, assigns in self.window_assigns.items():
+            local_fns = self.module_fns[module]
+            for prop, value in assigns:
+                if value not in local_fns:
+                    bad.append(f"  {module}: window.{prop} = {value}  "
+                               f"('{value}' not declared in {module})")
+        assert not bad, (
+            "Cross-file window.X = X assignments found (will cause ReferenceError):\n"
+            + "\n".join(bad)
+        )
+
+
+# =========================================================================
+# Group F — Module Size Sanity
+# =========================================================================
+
+class TestModuleSizes:
+    """No single module should be unreasonably large (regression guard)."""
+
+    MAX_LINES = 15000  # generous; largest module (wishlist-tools) is ~7200
+
+    @pytest.mark.parametrize("module", SPLIT_MODULES)
+    def test_module_size(self, module):
+        text = _read(_STATIC / module)
+        lines = text.count("\n") + 1
+        assert lines < self.MAX_LINES, (
+            f"{module} has {lines} lines (max {self.MAX_LINES})"
+        )
+
+    def test_total_lines_reasonable(self):
+        """Combined split modules should be in the same ballpark as the original."""
+        total = 0
+        for module in SPLIT_MODULES:
+            total += _read(_STATIC / module).count("\n") + 1
+        # The original was ~78K lines; allow 60K-100K for flexibility
+        assert 50000 < total < 120000, f"Total lines: {total}"